dtenedor commented on code in PR #40732:
URL: https://github.com/apache/spark/pull/40732#discussion_r1165912480
##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/ResolveDefaultColumns.scala:
##########
@@ -47,9 +47,11 @@ import org.apache.spark.sql.types._
* (1, 5)
* (4, 6)
*
- * @param catalog the catalog to use for looking up the schema of INSERT INTO
table objects.
+ * @param resolveRelation function to resolve relations from the catalog. This
should generally map
+ * to the 'resolveRelationOrTempView' method of the
ResolveRelations rule.
*/
-case class ResolveDefaultColumns(catalog: SessionCatalog) extends
Rule[LogicalPlan] {
+case class ResolveDefaultColumns(
+ resolveRelation: UnresolvedRelation => LogicalPlan) extends
Rule[LogicalPlan] {
Review Comment:
Good question, the answer is because the `ResolveRelations` rule is an
object that is nested inside the `Analyzer`. So we'd either need to pass a
reference to the current analyzer in order to do that, or pass a reference to
the `ResolveRelations` object as a whole (which is what I did before), or
otherwise pass a reference to the method of interest (which is what Gengliang
asked me to do instead). We don't anticipate any need to use a different
resolver anywhere, we just have to make the method visible.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
